416 PEOF. P. MAESHALL ON THE [Aug. 1906, 



of classification ; but this is the natural final goal, and present 

 attempts and descriptions should be largely based on those characters 

 that evidently are important from a genetic standpoint, although 

 their full significance is unknown. The suggested classification 

 seems to lose sight of these characters, and therefore neglect and 

 indifference to them would follow its general adoption, to the great 

 loss of petrography. 



A classification according to the petrographical or mineralogical 

 characters of the rock in each subrang would probably lead to 

 endless overlapping and confusion. In a word, it may be said that, 

 so far as the present group of rocks is concerned, the suggested 

 chemical classification, while grouping them according to their 

 chemical relationships, is unsatisfactory from a geological standpoint. 



The amount of care and thought that has been expended in 

 framing the system must have been almost colossal, and whether 

 the result is accepted as satisfactory or not, it must have a helpful 

 and salutary influence, of a marked nature, on the progress of 

 petrographical science. 



An attempt was made to discover any further possible relation- 

 ships between the rock-groups, by representing their chemical 

 composition by graphic methods. The diagrams, however, did 

 little more than figure forth the facts already mentioned : — 



(1) That the alkaline rocks show a graduated series, from the trachytes 



through the tiuguaites, to the least acid type of the trachytoid 

 phonolites. 



(2) That the basic rocks are also 'closely related one to the other. 



(3) That the trachydolerites and basanites form connecting-links, the 



former being closer to the alkaline series, the latter to the basic. 



It was considered that these facts were sufficiently evident, 

 without being further emphasized by the reproduction of the 

 diagrams. 



IV. Oeigin of the Diffeeent Kinds of Rocks. 



It is obvious that the occurrence of so many different kinds of 

 rocks in one district would, if interpreted correctly, throw con- 

 siderable light on the much discussed cause of the differences between 

 igneous rocks. 



So far as the succession of lavas is concerned, there are two 

 sections that display the actual relative position of lavas that seem 

 to be of prime importance. 



The first (fig. 1, p. 417) is exposed along the side of. the Water of 

 Leith from the corner of Duke Street and Royal Terrace to Skey's 

 House. Here the lowest rock is a basalt : a flow of Papanui Dolerite 

 50 feet thick rests upon the basalt, and above the dolerite comes an 

 andesite. This is succeeded by two flows of trachytoid phonolite, 

 above which comes a basalt, and finally the series in this district is 

 completed by a trachydolerite. 



